Re: HOT patch - version 15 - Mailing list pgsql-patches

From Pavan Deolasee
Subject Re: HOT patch - version 15
Date
Msg-id 2e78013d0709101215r5b92be68ifaa0fde92cafd3ee@mail.gmail.com
Whole thread Raw
In response to Re: HOT patch - version 15  ("Heikki Linnakangas" <heikki@enterprisedb.com>)
List pgsql-patches


On 9/10/07, Heikki Linnakangas <heikki@enterprisedb.com > wrote:

Oh, one more thing occured to me. Without HOT, we not only mark index
tuples pointing to dead tuples as killed, we remove them altogether if
the index page gets full. If you modify the test case so that after
doing the updates, you insert a bunch of tuples with a different key to
fill the index page, you should see CVS HEAD winning HOT without pruning
hands down.


Um. I am not sure I follow that. With HOT even if the HOT chain is long,
there is still just one index entry for the entire chain. So I don't see
why CVS HEAD would do better than HOT in the above case. Net-net
there will be equal number of index keys after the inserts.

Thanks,
Pavan

--
Pavan Deolasee
EnterpriseDB     http://www.enterprisedb.com

pgsql-patches by date:

Previous
From: "Pavan Deolasee"
Date:
Subject: Re: HOT patch - version 15
Next
From: Bruce Momjian
Date:
Subject: Re: HOT patch - version 15